Abstract
A 2-yr pitfall-trap study was conducted to examine the species composition of the Opiliones and the seasonal age-structure of Phalangium opilio (L.) in central Kentucky soybean, grassland, and alfalfa habitats. Phalangium opilio was the most commonly caught opilionid in all habitats, accounting for >90% of the adult specimens captured in soybean and alfalfa. Pitfall-trap results also suggest that Kentucky P. opilio populations have three generations per year and overwinter in the egg stage. Absolute sampling indicated that P. opilio was present at densities of <0.5 per m2 in soybean.
Members of the order Opiliones are found in many parts of the world and are often common where they occur. Many species are found in disturbed habitats, and some have been identified as predators in agricultural settings. The ecology of these organisms remains, for the most part, poorly known.
In a study of generalist predators in several Kentucky crops, Pfannenstiel (1995) found that "the Phalangidae" accounted for 13.4% (104 total observations) in 1993 and 17.6% (out of 74 total observations) in 1994 of recorded predation events on eggs of the corn earworm, Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae), in soybean fields during 24-h observation cycles. These percentages were exceeded only by predation events involving Nabis spp. (Hemiptera: Nabidae) and Geocoris punctipes (Say) (Hemiptera: Geocoridae). In another study, "Opiliones" were responsible for 39.0% (second only to Nabis spp.) of observed predation events (59 total observations) of H. zea eggs on Kentucky soybean on two dates in 1994 (Anderson 1996). And, from unpublished data gathered by K.V.Y. and W. E. Barney in Kentucky soybeans on three dates in 1993, "Opiliones" were responsible for 12.0% of observed predation events on H. zea eggs. It is not known which Opiliones species accounted for the predation events observed in these studies.
Several Opiliones species are known to occur in Kentucky, particularly Leiobunum spp. (Opiliones: Sclerosomatidae), but only a few studies (Bishop 1949, Bristowe 1949, Halaj and Cady 2000) have identified any of these species as predators in agricultural habitats. The harvestman Phalangium opilio L. (Opiliones: Phalangiidae) has been identified as a predator in a number of agroecosystems, including New Zealand strawberries, alfalfa, and cabbage, and potato fields in Scotland and Michigan (Ashby and Pottinger 1974, Leathwick and Winterbourne 1984, Butcher et al. 1988, Dixon and McKinlay 1989, Drummond et al. 1990). A well-known and cosmopolitan organism, P. opilio is common in Europe and North America (Savory 1962), and has been recorded from parts of Asia, Africa, New Zealand, and Japan (Ashby and Pottinger 1974, Edgar 1980, Suzuki and Tsurusaki 1983, Butcher et al. 1988). Although previously unreported in Kentucky, P. opilio has been found in a number of nearby states, including Ohio, Michigan, New York, Pennsylvania, Arkansas, Illinois, and Virginia (Cokendolpher and Lee 1993, Clark et al. 1994).
Pitfall traps are commonly used to gather information on the species composition, seasonal abundance, and population age-structure dynamics of ground dwelling arthropods (Kharboutli and Mack 1991, Wiedenmann et al. 1992), including opilionid faunae (Jennings et al. 1984, Drummond et al. 1990). We used pitfall traps from June 1996 through August 1998 to monitor the species composition of Opiliones and the phenology of P. opilio in soybean and nearby habitats. Because pitfall traps cannot be used to accurately measure population densities, we used a direct sampling method in the summers of 1997 and 1998 to estimate the absolute population density of Opiliones species in soybean.
Materials and Methods
Experimental Procedures.
The species composition of the Opiliones and the seasonal age structure of P. opilio were examined in soybean and two adjacent disturbed perennial habitats (alfalfa and grassland) using large-capacity pitfall traps based on a design developed by Houseweart et al. (1979) for long-term pitfall trapping, a design that was used by Jennings et al. (1984) in their study of Opiliones species composition and habitat preference in Maine spruce-fir forests.
For our study, each trap consisted of a 2-liter plastic bottle with the neck modified (about 2 cm cut from the top) to facilitate a plastic 15-cm-diameter powder funnel, the pouring spout of which was tightly fitted into the mouth of the bottle such that it could be removed from the bottle and then refitted each time the trap was checked and reset. Traps were placed in postholes (46 cm deep, 18 cm diameter), each of which was covered with a coated (two coats of polyurethane varnish) 30 by 30 by 0.5-cm hardboard apron, from which was cut a 15-cm-diameter hole such that the rim of the funnel was supported in the center of the apron. The preservative used in each trap was about 300 ml of a mixture of 1:1 ethylene glycol: 70% ethanol. Rain covers were not used, as shade-providing structures can influence sampling (Houseweart et al. 1979).
The traps were placed in two primary sites at the University of Kentucky's North Farm near Lexington. Each of the two sites was composed of three habitats, namely soybean, grassland (predominantly Kentucky bluegrass and fescue), and alfalfa, with three traps per habitat, for a total of 18 traps. Within each habitat at each site, traps were placed ≈5 m apart, and habitats within each site were ≈50 m apart. Soybeans (variety KS5292) were planted 20 May 1996, 12 May 1997, and 14 May 1998 and were not cultivated. Soybean plots were plowed 2-4 wk before planting. Established stands of grass and alfalfa were mowed three times per season, but otherwise were not disturbed. From June 1996 through August 1998, traps were run continuously and checked weekly during the growing season (March to November), and set up and monitored for 1-wk runs during each winter month (December, January, and February). Checking the traps consisted of removing the bottle from the posthole, separating it from the funnel, and pouring the contents through a fine sieve. The material in the sieve was then stored in 70% ethanol and taken to the laboratory. The strained liquid preservative was returned to the trap bottle and refreshed with unused preservative (the amount of which depended on the dilution and quantity of used preservative) such that at least 1/2 of the 300 ml in the trap bottle for the next week's run was fresh solution. In the laboratory, the arthropod contents from each trap were examined and all Opiliones individuals were removed.
Species Composition of the Opiliones in Soybean and Adjacent Disturbed Habitats.
Because there is no published method for identifying the instar of immature opilionids within the Phalangioidea, the superfamily to which all opilionid adults collected in our pitfall-trap samples belonged, all Opiliones specimens captured in pitfall traps were identified as being either "mature" or "immature" by the presence or absence of a genital opening beneath the operculum (Sankey and Savory 1974). Because no adult specimens were found that were neither P. opilio nor Leiobunum spp., and because very few Leiobunum individuals were captured, mature individuals were then identified as being either P. opilio or a Leiobunum sp. on the basis of several morphological characteristics. Phalangium opilio is distinguished from other Phalangioidea species known to occur in this area by the presence of a pair of tubercles on the supercheliceral lamella and by a lack of dentition on the pedipapal claw (Walker 1928, Bishop 1949, Sankey and Savory 1974).
The Population Age-Structure of P. opilio in Soybean Fields and Adjacent Disturbed Perennial Habitats.
Adult P. opilio specimens and immature Opiliones captured in pitfall traps were examined for seasonal age-distribution patterns. Because of the preponderance of P. opilio adults and the difficulty in identifying immature Opiliones to species, all immature Opiliones that superficially resembled P. opilio (only a few specimens with a distinct leg-banding pattern not known to occur on adult P. opilio were excluded) were considered in the age-structure analysis. Immature specimens were divided into three size classes, rather than by developmental instar, based on the width of the cephalothorax as measured just behind the third pair of coxae. These measurements were developed using laboratory-reared P. opilio specimens of known instar, where 10 of each immature P. opilio instar were measured. "Small" immature specimens were defined as having a cephalothorax width of less than 1 mm. These included first, second, most third, and some fourth instars, based on the colony-reared P. opilio example specimens. The cephalothorax width of "medium" individuals was between 1 and 1.5 mm, and these were equivalent to some third instars, most fourth instars and all fifth instars of colony-reared P. opilio. "Large" immature specimens had a cephalothorax width >1.5 mm. This group included sixth and seventh instars, based on colony-reared P. opilio. Voucher specimens are deposited at the University of Kentucky Department of Entomology Insect Collection.
Statistical Analysis.
For most comparisons, the effects of habitat type and sampling year on the numbers of P. opilio captured in pitfall traps were analyzed using one-way analysis of variance (ANOVA) and protected least significant difference tests on transformed [√ (x + ½)] mean total numbers of individual adult P. opilio captured in individual traps over a given time interval. All tables show untransformed values. For those cases in which transformations failed to correct heterogeneity of variance, a Kruskal-Wallis one-way ANOVA was used. The comparisons made using Kruskal-Wallis were: the effect of sampling year on the mean number of adult P. opilio trapped in alfalfa; the effect of habitat type on the mean amount of adult P. opilio captured from 17 June through 19 August 1996; the effect of sampling year on the mean number of adult P. opilio trapped in all habitats combined; and the effect of habitat type on mean numbers of adult P. opilio trapped for all sampling years combined. For those data analyzed with a Kruskal-Wallis test, multiple comparison methods described by Hollander and Wolfe (1973) were used to compare means. For all analyses, P ≤ 0.05 was the criterion used to determine significant effect.
Absolute Population Density of P. opilio in Soybean Fields.
The absolute population density of P. opilio in soybean was estimated by a direct sampling method. These samples were taken over three different dates in mid-July and early August during both 1997 and 1998 for a total of six sampling dates. On each date, two examiners sampled eight 2-m sections of two adjacent soybean rows between 2200 and 2400 hours EDT. These times were chosen based on observations by Pfannenstiel (1995), Anderson (1996) and others that many of the feeding events attributable to Opiliones species occurred during the first few hours after sunset.
Using two large (2 m by 1 m by 0.5 cm, length by height by thickness) hardboard sheets that were painted white to aid in illumination and vision, the two examiners isolated one of the randomly predetermined sample sites by carefully setting the hardboard sheets down vertically midway between the outside of the sampling area and adjacent rows in a way such that the boards were parallel with the rows. Each board was fixed with two struts on the side facing away from the sampling area, which allowed the boards to stand perpendicular to the ground while sampling occurred. These boards were designed to help control movement of arthropods into and out of the sampling area, and also to isolate the sampling area from overhanging soybean plants in nearby rows. When the boards were in place, the two investigators, wearing headlamps, entered the sampling area from opposite ends between the boards and, working toward each other, proceeded to beat the soybean plants and dislodge arthropods onto the ground between the two rows. Any dislodged Opiliones specimens were collected. The ground and leaf litter in the sampling area were then thoroughly examined for remaining Opiliones specimens. Collected specimens were preserved in 70% ethanol and taken to the laboratory for identification.
Results and Discussion
Species Composition of the Opiliones in Soybean Fields and Adjacent Disturbed Habitats.
Because of the difficulty in identifying immature Opiliones, the species composition of pitfall-trap samples was addressed by comparing numbers of captured adult specimens only. For the whole trapping period, P. opilio accounted for >70% of adult specimens captured in all three habitats, and >95% of adult specimens captured in soybean (Table 1). The remainder of collected adult individuals was Leiobunum spp. These numbers suggest that P. opilio is the most common opilionid in soybean, grassland, and alfalfa habitats in central Kentucky.
Total number and percentage of individual adult Opiliones specimens, by taxa, captured within each habitat, 17 June 1996-18 August 1998
Total number and percentage of individual adult Opiliones specimens, by taxa, captured within each habitat, 17 June 1996-18 August 1998
During the winters of 1996-1997 and 1997-1998 trapping periods, only a small number of adult P. opilio, and no other adult or immature specimens, were captured in December. No specimens were captured in January or February. Numbers of adult P. opilio captured during the two winters combined were 14 in soybean, nine in grassland, and 17 in alfalfa.
The Population Age-Structure of P. opilio in Soybean Fields and Adjacent Disturbed Perennial Habitats.
The immature Opiliones and adult P. opilio collected in each of the three habitats provide insight concerning the probable voltinism and overwintering stages of P. opilio. Specimens collected in pitfall traps placed in alfalfa from March to November 1997 in particular show distinct peaks of activity for the three age classes, with the peak of each age class occurring just a few weeks after the peak of the immediately smaller age class (Fig. 1). Although most of the "small," "medium," and "large" nymphs represented in these figures are believed to be P. opilio, based on the large numbers of adult P. opilio captured in these habitats, the specific identities of individual specimens remain unknown. The relationships between peak occurrences of "large" nymphs and adult P. opilio, however, lend further support to the hypothesis that these three classes of captured immature Opiliones are comprised of mostly P. opilio.
Mean numbers of individual immature Phalangioidea and adult P. opilio (±SE, n = 6 traps) trapped per week in alfalfa, 17 June 1996-18 August 1998.
Assuming the three immature age classes are comprised mainly of P. opilio, the trends represented by these peaks also suggest that P. opilio undergoes three generations each year in our area, with a first generation of early instars active in late March through early May, leading to a peak of adult activity in late May through mid-June. Another peak of early instar activity follows in June and July with adults active in July and August. A third peak of early instar activity appears in August and September, with a less pronounced period of adult activity in October and November. The apparent decline in adult activity late in the year, coupled with the complete absence of adults and presence of small nymphs during early spring, suggests that eggs are the predominant overwintering stage for P. opilio in Kentucky.
These results differ somewhat from the phenologies reported for P. opilio in other geographic regions. German populations of P. opilio are believed to be univoltine, with immature individuals emerging in the spring and maturing throughout summer, and with mating and oviposition occurring in autumn followed by an overwintering of the egg stage (Bachmann and Schaefer 1983). British P. opilio populations were found to follow a pattern similar to that reported in Germany, but, "at least in favorable southern conditions," overwintering may occur among both eggs and immature specimens, thus producing two generations the following year (Sankey and Savory 1974). Bishop (1949) stated, in reference to New York P. opilio populations, that "adults of P. opilio will survive a mild winter," and Edgar and Yuan (1968) reported that Michigan P. opilio have "two or three generations per year" and that they overwinter as "eggs, young, and probably adults." Drummond et al. (1990) used pitfall traps to monitor ground-predator activity in Michigan potato fields from 1 July to 18 September 1986 and 23 May to 7 September 1987 and found that adult P. opilio were active during the early part of both sampling seasons until early to mid-August, when the numbers of captured adults dropped dramatically and the number of immature P. opilio increased and remained active until the end of the sampling periods. Pitfall traps also were used by Martin (1983) from August 1971 to June 1973 to monitor the ground fauna of New Zealand pastureland. That study showed two generations for P. opilio in the trapping area, with adults first appearing in early spring followed by juveniles, with another wave of juveniles appearing in summer. In addition, "the presence of small juveniles in April and larger juveniles and males and females in May and June suggested a third generation." In a New Zealand pitfall-trap study designed to monitor predatory arthropods in carrot fields, Sivasubramaniam et al. (1997) reported that P. opilio was caught in traps "throughout the season" (November-March) "with main activity in January and again in March." No information was presented in reports of any of these phenological studies concerning the means used to identify immature opilionids to species. The studies by Martin (1983) and Sivasubramaniam et al. (1997) were designed to investigate and report species composition, and in both of those studies P. opilio was the only Opiliones species mentioned, suggesting it was the only adult species found. If that was the case, it is probable that the immature opilionids sampled in those studies were also P. opilio.
The differences in number of generations per year and suspected overwintering stages are likely due to population or climatic differences between regions. It is also possible that there may be differences in the seasonal activity of P. opilio based on habitat. For instance, adults appeared in early spring in New Zealand (Martin 1983), but not in our study. Perhaps adults overwinter in our area also, but are not present in alfalfa, grassland, or soybean habitats until later in the year.
Although similar peaks of activity-density are suggested by our other results, particularly in the data for March to November 1997 in soybean (Fig. 2), they are most clearly illustrated in the pitfall-trap data for the alfalfa habitat during 1997 (Fig. 1). This is due to the relatively large numbers of individual opilionids caught in alfalfa during this period compared with the other two habitats and the other 2 yr of sampling in alfalfa. The mean number of adult P. opilio captured from mid-June to mid-August in 1997 in alfalfa was more than twice the mean number captured in 1996 or 1998 in alfalfa, and in the other two habitats during this period in any of the 3 yr (Table 2). This was a significantly greater number of adult P. opilio captured (F = 22.43; df = 2, 15; P < 0.01) when compared with the mean numbers captured in the other two habitats during this period in 1997, and significantly more P. opilio on average (F = 62.91; df = 2, 15; P < 0.01) were captured during this period in 1997 than during this period in 1996 or 1998 in alfalfa. In addition, when comparing among years for the trapping period of mid-June to mid-August for all habitats combined, the mean number of adult P. opilio captured in 1997 was significantly greater (F = 20.91; df = 2, 51; P < 0.01) than the mean captured in 1998, and twice the mean number captured in 1996 (although this difference was not significant). Similarly, the mean number of P. opilio captured for all years combined was significantly greater (F = 7.22; df = 2, 51; P < 0.01) in alfalfa traps than in grass traps. Although there was a trend for more adult P. opilio to be captured in soybean than grass, the mean number captured in soybean for all years combined was not significantly different from the mean number captured in grass. When looking at the mean number captured in each year for each habitat, there was also a lack of significant difference between the mean numbers captured in soybean and grass during the 1997 and 1996 (Table 2), but during 1998 the mean number captured in soybean was significantly greater than the number captured in grass or alfalfa (F = 6.50; df = 2, 15; P = 0.01). The trapping period for 1998, the period during which the fewest adult P. opilio were captured in all habitats, was the only period in which the mean number of adult P. opilio captured in soybean was significantly greater than the means captured in the other two habitats. The average number caught per trap in soybean (10.00 ± 2.97) was low during this period in 1998, but this average was not significantly different (F = 3.03; df = 2, 15; P = 0.07) from the average number caught in soybean during the same period in both 1996 and 1997.
Mean numbers of individual immature Phalangioidea and adult P. opilio (±SE, n = 6 traps) trapped per week in soybean, 17 June-18 August 1998. Numbers adjacent to the truncated error bars indicate that an addition of the given value represents the true standard error value.
Mean numbers of adult P. opilio captured per pitfall trap in three habitats during three common trapping periods
Means within the same column followed by the same letter (a, b, or c) or within the same row followed by the same letter (x or y) do not differ significantly (F > 0.05).
Mean ± SE, n = 6 traps.
Mean ± SE, n = 18 traps.
Mean numbers of adult P. opilio captured per pitfall trap in three habitats during three common trapping periods
Means within the same column followed by the same letter (a, b, or c) or within the same row followed by the same letter (x or y) do not differ significantly (F > 0.05).
Mean ± SE, n = 6 traps.
Mean ± SE, n = 18 traps.
Absolute Population Density of P. opilio in Soybean Fields.
Few Opiliones specimens were captured in the absolute population density sampling study. During 1997, an average of 0.15 adult P. opilio specimens per m2 were captured. Several immature Opiliones were captured on the three sampling dates in 1997, and if these are assumed to be immature P. opilio (which they all superficially resembled), the average number of suspected P. opilio collected, including adults, becomes 0.38 individuals per m2. For all three sampling dates in 1997, only one individual, a Leiobunum sp. adult, was captured that was not suspected to be P. opilio. In 1998, the average number of adult P. opilio captured was 0.08 per m2. This average increases to 0.13 individuals per m2 when including the immature specimens, all of which resembled P. opilio. Two individual adult Leiobunum sp. individuals were collected during 1998. For both years combined, the average number of adult P. opilio collected was 0.11 per m2. The average number of suspected P. opilio (adult and immature) for both years combined was 0.25 per m2.
These numbers suggest that the population density of P. opilio in soybean during late July and August is low, despite pitfall-trap results (Fig. 2), which suggest that this period includes the peak of adult and late instar P. opilio activity in soybean. Considering the relatively large numbers of P. opilio captured in pitfall traps in soybean, the low absolute densities indicated by this study seem incongruous. The numbers of Opiliones captured in pitfall traps, however, represented the accumulation of an entire week of activity. If the traps were passive collectors, then the relatively low absolute densities suggest that P. opilio is highly active on the ground.
In summary, the results of our pitfall-trap study indicate that P. opilio is the most common Opiliones species in soybean and alfalfa fields, as well as nearby grassland, in central Kentucky, and is present in those ecosystems for a large part of the growing season. This is consistent with previous research in other agroecosystems in various parts of the world and supports the pattern of P. opilio being able to survive and reproduce in disturbed habitats in temperate climates. Phalangium opilio appears to have three generations per year in Kentucky and probably overwinters in the egg stage, which contrasts with reports from other geographic regions. In several of those studies, adults and juveniles, as well as eggs, were suggested as possible overwintering stages, and fewer than three generations per year were indicated, particularly at higher latitudes. It appears that P. opilio is capable of adjusting its life history to the specific climates of many different regions. Finally, our results from pitfall-trap sampling and absolute sampling, combined with the relatively high frequency of predation by Opiliones on H. zea eggs observed by Pfannenstiel (1995) and others, suggest that P. opilio are highly mobile predators.
A previous draft of the manuscript was critically reviewed by D. L. Dahlman and S. L. Dobson (University of Kentucky). The investigation reported in this article (00-80-205) is in connection with a project of the Kentucky Agricultural Experiment Station.
References
